Patient Satisfaction in Phlebotomy

Transcription

1 PHLEBOTOMY JaneC. Dale, MD Peter J. Howanitz, MD Patient Satisfaction in Phlebotomy A College of American Pathologists' Q-Probes Study From the Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minn (Dr Dale), and the Department of Pathology and Laboratory Medicine, University of California, Los Angeles Medical Center (Dr Howanitz). Reprint requests to Dr Dale, Department of Laboratory Medicine and Pathology, Hilton 78, Mayo Clinic, First St SW, Rochester, MN 9. The phlebotomy service is the most common, and often the only, segment of the laboratory involved in direct patient contact. The quality of that contact, which requires both technical and interpersonal skills, is reflected in the patient's level of satisfaction with the experience. The Joint Commission on Accreditation of Healthcare Organizations (JCAHO) recognizes patient satisfaction as an important indicator of quality and requires its members to collect data about expectations of patients and the degree to which those expectations have been met.1 The College of American Pathologists' (CAP) Q-Probes program is a national, voluntary quality improvement program that develops and uses standardized tools to gather data, assess practices, determine performance, and describe opportunities for improvement in a wide range of laboratory activities. Since its inception in 1989, the program has evaluated quality indicators related to all areas of the laboratory testing process. 8 LABORATORY MEDICINE VOLUME 7, NUM BER on 8 May1 818 day and patients undergoing arterial punctures, capillary punctures, or indwelling line draws. We obtained demographic information from yearly Q-Probes subscriber demographic forms. Participating laboratories were supplied with postcards with questions about the patient's phlebotomy experience. Using one card per patient, laboratory personnel recorded the date and time the patient arrived at the laboratory for the venipuncture, the time the procedure was initiated, and the number of venipunctures required to obtain the appropriate specimen(s). Patients were instructed to complete the postcard on the following day and to return the postcard to the laboratory. Patients were asked to measure and record the size of the bruise diameter at its widest point using the ruler printed on the postcard, and to indicate, with a check mark in the appropriate box, the following: whether they were taking blood thinners, steroids, or chemotherapy; whether the waiting time was the same, longer, or shorter than they expected; whether any discomfort experienced was the same, more, or less than they expected; whether they were treated in a courteous, professional manner; whether they were satisfied with the procedure; and whether Materials and Methods ABSTRACT We monitored satisfaction with phlebotomy In the first six months of 1994, CAP's Small procedures in 4,96 outpatients from 9 participating Hospital Q-Probes program undertook a study of institutions. More than 99% of patients were satisfied with patient satisfaction with laboratory phlebotomy the phlebotomy services provided. The factor that correlated services. This study was designed to assess outpatient satisfaction with phlebotomy services and to most highly with patient satisfaction was courteous and pro- relate patient satisfaction to key aspects of qualifessional treatment. Other factors, which showed a much ty, including waiting time, number of needleweaker correlation with patient satisfaction, included, in sticks required, presence of bruises, discomfort, order of decreasing correlation: the absence of large bruises,and the patient's perception about whether he or the need for only one needlestick to obtain an appropriate she was treated in a courteous and professional manner. Outpatients who presented to particispecimen, patient recognition of an outstanding employee pating laboratories for phlebotomy were invited during the phlebotomy procedure, less patient discomfort to enroll in the study. Excluded from the study than anticipated, and waiting time of less than minutes. were patients scheduled for surgery the following

2 Results Three hundred ninety-three institutions participated in the study. The analysis of these participants by institution type and average occupied bedsize is shown in Table 1. The average occupied bedsize ranged from to 8 beds, with a mean of 68 occupied beds per institution. The median number of pathologist and phlebotomist fulltime equivalent (FTE) positions per institution was one and three, respectively. Enrolled patients ranged in age from less than 1 to 96 years, with an average of 6 years. Of 4,96 postcards distributed to patients, 1,814 were returned for an overall response rate of 64%. The mean number of postcards distributed was 89 per institution, and the range was from 1 to Vicky Culver, 16. Fifty-three percent of participating laboratories MT(ASCP), right, distributed all 1 postcards; 1.% distributed more draws blood from Hannah Heinritz, than 1 postcards. The percentile distribution of while Priscilla Rozek, performance variables, including patient response MT(ASCP), comforts rate, among participating institutions is shown in the child. Table. There was no indication that institutions with higher response rates also had greater percentages of dissatisfied patients, and vice versa. The average waiting time in participating institutions, calculated as the difference between time of registration and time of phlebotomy initiation, was 8.4 mintable 1. PARTICIPANT utes. Laboratory perdemographics sonnel recorded these two times. The perparticipants Characteristic centile distribution of median waiting times Institution Type (n = 9) is shown in Table. 9.9 Hospital For example, the 9th Clinic or independent percentile indicates laboratory 1. that 1% of particilong-term care facility pants had median. waiting times of Other 4. minutes or less. Only Average No. of 49.% of patients indioccupied Beds (n = 7) cated the wait was. shorter than expected, while.8% of patients indicated the wait was longer than expected. The average waiting time for those patients. 11- who indicated they > 1.9 expected a shorter wait was 16.1 minutes. For those patients who indicated they expected a longer wait, the average waiting time was 7. minutes. The mean waiting VOLUME 7, NUMBER LABORATORY MEDICINE 189 they had met an outstanding employee. Patients also listed their age and their estimated waiting time prior to having their blood drawn. If patients were unable to complete the phlebotomy card at home, cards completed by others were accepted. Each multiple-choice question on the phlebotomy card could be answered as "not known." The study was conducted for months or until 1 postcards were distributed by each laboratory, whichever occurred first. The laboratory transferred information about patients to standardized input forms and submitted the forms to CAP for data handling. For those institutions that enrolled fewer or more than the 1 patients required for the study, the interinstitutional ranking reflected responses received per number of patients enrolled. Participating laboratories also answered 18 questions about the demographics, policies, and procedures of their phlebotomy service. Four questions required numeric answers, nine were yes or no, and five were best-answer multiple choice with two to five choices. Four of the multiplechoice questions were answered only if a previous yes or no question was answered with a yes. If participants failed to answer a question, they were excluded from the database for that question only. For percentile rankings of participating laboratories for each performance variable, a higher percentile ranking was associated with better performance. Statistical significance of quantitative variables was evaluated by the Wilcoxon's rank sum test. The association between performance variables and satisfaction was tested by the Pearson x statistic. We defined as statistically significant a P value equal to or less than.. We compared these results from the small hospital setting with those from a similar study conducted in a larger hospital setting.

3 TABLE. DISTRIBUTION <OF PERFORMANCE VARIABLES AMONG PARTICIPATING 1INSTITUTIONS Percentile* th (Median) 9th Patient response rate Median patient waiting time (minutes) Patients satisfied with procedure Bruise diameter (mm)' I * The higher percentiles correspond with better performance. f If a patient had more than one bruise, the sum of diameters was tised. TABLE. REPORTED PHLEBOTOMY CHARACTERISTICS Number of Attempts per Patient Patients time for patients who i n d i c a t e d they were satisfied w i t h t h e i r phlebotomy experience was 8.4 minutes, c o m p a r e d with 1.4 m i n u t e s for p a t i e n t s who were dissatisfied. Ninety-seven per- 1 ".:. Number of Bruises per Patient cent of blood collec t i o n s r e q u i r e d only one venipuncture (Table ). These results were similar to those...? of a previous Q-Probes Degree of Discomfort Reported s t u d y of i n p a t i e n t More than expected 6 phlebotomy practices, Less than expected 7 which showed that The same as expected % of successful p h l e b o t o m y p r o c e d u r e s r e q u i r e d only one attempt. 4 Laboratories with a higher percentage of phlebotomies requiring only one venipuncture had a greater percentage of satisfied patients than laboratories with a lower percentage of phlebotomies requiring one venipuncture. Approximately 87% of patients reported no bruising related to the phlebotomy procedure, and 1.9% noted the presence of one or more bruises (Table ). For all bruises recorded, the average diameter was 1. mm. When more than one bruise was recorded, the bruise size was determined by the sum of individual diameters recorded. For the,797 patients who recorded bruise diameter, % of bruises were less than m m, % were between and 1 m m, % between 1 and mm, and another % were greater t h a n m m. Patients who were not bruised were more likely to mention an outstanding employee and reported a higher percentage of 1 9 LABORATORY MEDICINE VOLUME 7, NUMBER Each patient was asked several questions related to his or her level of satisfaction with the phlebotomy experience. Their responses indicated that 99.7% felt they were treated in a courteous and professional manner, 99.% were satisfied with the procedure, and 9.4% had encountered an employee they characterized as "outstanding" during the procedure. The distribution of participants' patient satisfaction rates (percentage of patients who were satisfied with the p h l e b o t o m y p r o c e d u r e ) is shown in Table. Sixty-nine percent of participating laboratories had satisfaction rates of 1%. Information provided by participants concerning phlebotomy practices and policies at their institution is provided in Table 4. None of these policies demonstrated a correlation with patient satisfaction. Factors associated with patient dissatisfaction are shown in Table. Discourteous treatment showed the strongest correlation. Other factors that showed weaker, but statistically significant correlations with dissatisfaction were, in decreasing strength of correlation, a large (> mm) bruise, more than one needlestick, failure to indicate the presence of an outstanding employee, more discomfort than expected, greater than minutes waiting time, and presence of a bruise of any size. A similar Q-Probes study conducted in 199 is compared with the current study in Table 6. Whereas the earlier study was aimed at large institutions, the current study was directed at institutions of less than beds. Fewer patients returned postcards in the current study than in the earlier study (64.% vs 7.7%); however, more patients reported meeting an outstanding employee in the current study than in the earlier study (9.4% vs 47.7%). 1th Variable single-attempt venipunctures than did patients who were bruised. Twenty-six percent of all patients indicated they were taking medications that could increase the likelihood of venipuncture-related bleeding. These patients had a slightly elevated rate of bruising (1.%), compared with patients who did not report taking these types of medications (1%). On average, patients who were taking these medications had slightly smaller bruises (1 mm) than did patients not taking these medications (16 mm). The differences in the number of bruises and in bruise diameter between these two groups of patients were statistically significant (P<.1). In addition, 6% of patients experienced more discomfort than anticipated (Table ).

4 A number of studies of patient satisfaction have shown that interpersonal aspects of care correlate highly with satisfaction and often assume more importance than technical aspects of care. In a meta-analysis of 1 studies of patient satisfaction, Hall and Dornan showed that humanness (defined as warmth, respect, kindness, willingness to listen, appropriate nonverbal behaviors, and interpersonal skill) ranked second out of 11 aspects of medical care associated with patient satisfaction, including cost, outcome, and facilities.6 Only overall quality showed a higher correlation with patient satisfaction. Cleary and McNeil found that "good communication skills, empathy, and caring appear to be the strongest predictors of how a patient will evaluate the care received." 7 In our study, discourteous treatment showed the strongest correlation with patients' dissatisfaction with the phlebotomy service. A number of patient characteristics are thought to be related to patient satisfaction, including sociodemographic characteristics (eg, age, social class, race, and gender), physical and psychological health, attitudes, and expectations.7 Our study did not find a correlation between age and satisfaction. Our patient population was not evenly distributed, however. Forty-five percent of patients fell into the 6- to 8-year-old age bracket. We did not include other sociodemographic characteristics in our study design, and failure to control for those characteristics may have introduced bias into our results. Nowhere in the laboratory are interpersonal skills more important than in the phlebotomy service. Phlebotomists not only must be skilled technicians, they also must have and demonstrate the attributes of humanness described above in TABLE 4. PHLEBOTOMY SERVICE PRACTICES AND POLICIES AT PARTICIPATING INSTITUTIONS Participants Practice/Policy Phlebotomists perform other tasks 9 Policy on limiting number of needlesticks 89 Patient relations course 6 Formal complaint procedure 8 Patient satisfaction monitored Waiting time monitored 8 Phlebotomy training course 17 Policy requiring patient to be seated prior to phlebotomy TABLE. TREATMENT FACTORS ASSOCIATED WITH PATIENT DISSATISFACTION Patients Patients Reporting Reporting Factor Who Were Treatment Factor Factor (Number) Dissatisfied Discourteous treatment Large bruise (> mm) More than one needlestick 6 7. No outstanding employee identified 1, 6.4 More discomfort than expected 1, ,8.4 In the business community, customer satisfaction long has been recognized as vital to the success of an organization. Increasingly, the health care industry is realizing the importance of patient satisfaction, not only as a means to ensure the institution's survival in this age of competition for patients but also as a reflection of quality. While it may be difficult for patients to accurately assess the quality of many hospital services, their level of satisfaction with the phlebotomy experience often is a reliable indicator of the quality of those services. The convenience of the service, the time involved, the number of needlesticks required, the occurrence of pain or bruising, and the attitude of the phlebotomist all relate to the quality of the phlebotomy service and can impact patient satisfaction. VI c Waiting time > minutes Any bruise nic. atio Discussion even the most difficult patient encounters. Often, the patient will judge the quality of the service he or she receives by the interpersonal skills of the individual who provides the service. This is especially true in phlebotomy. In a nationwide survey of more than 7, patients from 14 hospitals, a patient's perception of the courtesy of the technician correlated highly with the perception of the skill of the technician who collected the patient's blood.8 Our studies corroborate those findings. Information from patients' postcards correlated with their satisfaction with the phlebotomy service is shown in Table. Of the factors studied, courteous treatment showed the strongest correlation with satisfaction. Although only 61 patients (.%) felt they were not treated courteously, 49.% of those patients indicated they were not satisfied with the service they received. Other factors also showed an association with dissatisfaction that was statistically significant, but with VOLUME 7, NUMBER LABORATORY MEDICINE E E o c I ID

5 TABLE 6. COMPARISON OF RESULTS OF 199 AND 1994 PHLEBOTOMY Q-PROBES STUDIES Number of participating institutions 6 9 Median bed size ,7 4, Patients reporting bruising from phlebotomy procedure Mean bruise size (mm) reported by patients Patients who were satisfied with phlebotomy procedure Patients who reported meeting an outstanding employee Number of patients enrolled in study Percentage of postcards returned by patients (response rate) Length of study period (months) Patients reporting more discomfort than expected from phlebotomy procedure a much weaker correlation than that seen with discourteous treatment. These factors included, in order of decreasing correlation: presence of a large (> mm) bruise, two or more needlesticks, the absence of an outstanding employee, more discomfort than expected, a waiting time of greater than minutes, and the presence of a bruise of any size. An earlier Q-Probes study that was targeted primarily at hospitals greater than beds looked at complications of phlebotomy and addressed similar issues as did the present study. 4 Comparative results of both studies are shown in Table 6. Waiting time for patients is not included owing to differences in determining waiting time in each study. The patient satisfaction rate was slightly higher in this Small Hospital Q-Probes study than in the 199 study conducted in larger hospitals. The percentage of patients who met an outstanding employee was much higher than in 199. Other studies have shown that patients in smaller institutions and smaller communities tend to be more satisfied than patients in larger ones.'7 Waiting time was a factor only when it was longer than minutes. Five out of seven other factors were more important than a long waiting time. Factors directly associated with the phle- 1 9 LABORATORY MEDICINE VOLUME 7, NUMBER References 1. Joint Commission on Accreditation of Healthcare Organizations (JCAHO). Accreditation Manual for Hospital Standards. Oakbrook Terrace, 111: JCAHO; 1996:9.. Howanitz PJ, Howanitz JH, Wenzel RP, eds. The Clinical Laboratory: Assessing Quality Healthcare Perspectives for Clinicians. Baltimore, Md: Williams and Wilkins; 199: Howanitz PJ, Cembrowski GS, Bachner P. Laboratory phlebotomy: a College of American Pathologists Q-Probes study of patient satisfaction and complications in,78 patients. Arch Pathol Lab Med. 1991;11: Howanitz PJ, Schifman RB. Inpatient phlebotomy practices: a College of American Pathologists Q-Probes quality improvement study of,1,64 phlebotomy requests. Arch Pathol Lab Med. 1994;118: Press I, Ganey RF, Malone MP. Patient satisfaction: where does it fit in the quality picture? Trustee. 199;4:8-1, Hall JA, Dornan MC. What patients like about their medical care and how often they are asked: a meta-analysis of the satisfaction literature. SocSciMed. 1988;7: Cleary PD, McNeil BJ. Patient satisfaction as an indicator of quality care. Inquiry. 1988;: Press I, Ganey RF, Malone MP. Satisfied patients can spell financial well-being. Healthcare Financial Management. 1991;4:4-4. Result botomy procedure and how patients perceived they were treated had a greater influence on satisfaction than did waiting time. Despite the fact that waiting time did not have a significant effect on satisfaction until it was longer than minutes, it may be helpful to inform each patient of the anticipated waiting time whenever possible. Less than % of participants had a laboratory policy that required patients to sit for a specified period of time prior to phlebotomy. Perhaps the fact that patient satisfaction was not adversely affected until the waiting time was longer than minutes will encourage laboratories to adopt a minimum patient waiting time policy as a means of reducing preanalytical variability due to postural changes. As a measure of quality, patient satisfaction is easy to understand and relatively easy and inexpensive to measure. Because most patient complaints go unreported, those that are received may represent only the "tip of the iceberg" and should be investigated promptly and seriously. Periodic surveys of patients can help identify problems with patient care and can provide key information concerning patients' level of satisfaction with the services provided. This study shows that treating patients in a courteous and professional manner can have a profound impact on their level of satisfaction.